UNITED STATES COURT OF APPEALS FOR THE SEVENTH CIRCUIT
decided: August 28, 1974.
UNITED STATES OF AMERICA, PLAINTIFF-APPELLEE,
EWIG BROS. CO., INC., A CORPORATION, AND EUGENE W. EWIG, AN INDIVIDUAL, DEFENDANTS-APPELLANTS. UNITED STATES OF AMERICA, PLAINTIFF-APPELLANT, V. VITA FOOD PRODUCTS OF ILLINOIS, INC., A CORPORATION, AND LAWRENCE T. SCHWEIG, AN INDIVIDUAL, DEFENDANTS-APPELLEES
Appeal from the United States District Court for the Eastern District of Wisconsin. No. 71-C-158. MYRON L. GORDON, Judge. Appeal from the United States District Court for the Northern District of Illinois, Eastern Division. No. 70-C-2246. JAMES BENTON PARSONS, Judge.
Hastings, Senior Circuit Judge, and Stevens and Sprecher, Circuit Judges.
STEVENS, Circuit Judge.
There are two ways to state the principal question presented by these appeals. Narrowly, the issue is whether residues of DDT and dieldrin smoked chubs are "food additives" within the meaning of § 201(s) of the Federal Food, Drug and Cosmetic Act.*fn1 A somewhat more disturbing way to state the same question is whether all of the fish in the Great Lakes are "adulterated" as a matter of statutory definition.*fn2 If they are, the Administrator*fn3 may have, at least for the present, virtually unbridled power to eliminate all such fish from our food supply. We therefore attach special importance to the additional questions presented in the Vita Food appeal. That appeal, unlike the Ewig Bros. appeal,*fn4 requires us to consider the legal significance of an "interim guideline" announced in a press release on April 22, 1969, as well as the district court's findings that the testing methods used by the government's experts were not sufficiently reliable to demonstrate that Vita's smoked chubs contained more DDT than the guideline permits.
A total ban on the future use of DDT would not resolve the problem presented by this case. Although the levels of DDT contamination are declining, we must assume that the chemical, or its derivatives, will survive as an ingredient of all or most foods for some time.*fn5
Scientists seem to agree that if the DDT level is high enough, the food should not be consumed by man and, conversely, if the amount is sufficiently small, ingestion of DDT may be harmless. Danger levels have not been precisely defined. The record demonstrates, however, that in fish levels in the range of 5 parts per million are neither (a) generally recognized among qualified experts as safe,*fn6 nor (b) demonstrably injurious to health or unfit for human consumption.*fn7 At the levels disclosed by the record before us, the effect on human health is somewhat uncertain.
Unquestionably DDT is a "pesticide chemical" as that term is defined in § 201(q) of F.D.C.A. See 21 U.S.C. § 321(q). Pursuant to statutory procedures, tolerances have been established for its use in or on various raw agricultural commodities, including fruits, vegetables, and meat.*fn8 Such foods may therefore contain DDT within the prescribed tolerance limits without being "adulterated." It does not follow, however, that a level which is either safe or unsafe for one food would be equally safe or unsafe for another food.*fn9 With respect to the foods for which DDT tolerances have been set, it is reasonable to infer that the rule-making process may have been invoked either by the FDA itself, by manufacturers or distributors of the pesticide, or by farmers or producers who are interested in using DDT as a pesticide. Fishermen, however, have never had any interest in using or selling DDT themselves; its presence in the environment is a condition of their work -- and also of the business of distributing or processing fish -- for which they are not responsible and which they have no interest whatsoever in perpetuating. In short, unlike farmers and pesticide salesmen, they have never had any interest in adding DDT to the environment or to the food supply. From their point of view, it is not an item which is added to their products; it is a natural component of the fish before it is caught, let alone processed.
In this case the government's claim that defendants' chubs are "adulterated" is not predicated on a claim that the particular fish defendants sell contain a poisonous substance or are otherwise unfit for food pursuant to either subparagraph (1) or subparagraph (3) of § 402 of the F.D.C.A.*fn10 Under those subparagraphs the government would have the burden of proving that the fish are actually harmful to man. Instead, the government's claim is predicated on § 402(a)(2)(C),*fn11 under which it need only prove that "such substance is not generally recognized, among experts qualified by scientific training and experience to evaluate its safety, as having been adequately shown through scientific procedures . . . to be safe under the conditions of its intended use . . . ." § 201(s) of the F.D.C.A., 21 U.S.C. § 321(s).*fn12
It is the government's position that a fair analysis of the statutory scheme Congress has enacted, including the allocation of decision-making responsibility between the agency and the judiciary, justifies proceeding under this section. For if, as the government contends, DDT is a food additive, the Food and Drug Administration may itself decide when products containing quantities of DDT should be removed from public consumption, without having to rely upon the decisions -- possibly inconsistent with one another -- of different federal judges determining danger to health under §§ (a)(1) and (a)(3) on a case-by-case basis.
The question, then, is whether DDT and dieldrin in defendants' processed fish are "food additives" within the meaning of § 201(s).
We have recently identified the principal purposes of the food additive amendment of 1958.*fn13 Neither the purpose to establish a procedure for premarketing clearance of untested food additives, nor the intent to permit the evaluation of such products to encompass a consideration of their benefits, as well as their potential for harm, seems directly relevant to the question before us. Rather, we are concerned with products which, at least primarily, were intended to be regulated as pesticide chemicals rather than as food additives. It is therefore appropriate to review the development of the pesticide chemical legislation enacted in 1954 as well as the food additive amendment itself.
Prior to 1938 a poisonous substance added to a food during processing would not cause adulteration unless the government could prove that the food itself, as opposed to the added substance, was unsafe.*fn14 After 1938, however, the statute has focused attention on the character of the added substance. Section 406 of the 1938 Act prohibited the use of poisonous or deleterious substances unless industry demonstrated that the substance was required for processing and also persuaded the Administrator*fn15 to issue a tolerance regulation permitting use in amounts below a specified level. Pending the promulgation of such a tolerance regulation, the addition of any poisonous or deleterious substance -- even traces which could not possibly affect the safe character of the end product -- constituted adulteration as a matter of law.*fn16 The "per se" approach adopted in 1938 placed the burden of proving safety upon industry and, as a matter of procedure, mandated marketing delays until after formal administrative review could be completed.
Although the 1938 Act authorized the promulgation of regulations limiting the amount of pesticide residues that remain in or on food (since such chemicals were obviously "poisonous or deleterious"), only one such regulation was actually issued during the subsequent 15-year period. The absence of such formal tolerance regulations was attributable, in part, to the expense and complexity of the rule-making procedure and, in part, to the required showing that use of the added substance be essential or unavoidable during processing.*fn17 Instead of employing the statutory procedure, the F.D.A. exercised practical control by means of an extra-legal system of unofficial and informal tolerances.*fn18
The pesticide chemical amendment of 1954 was adopted to enable industry to make effective use of these poisonous products, even when not absolutely necessary, and to simplify, expedite, and improve the tolerance rule-making procedure*fn19 -- as well as to remove uncertainty from the law.*fn20
The 1954 Act created a category of added poisonous substances known as "pesticide chemicals" and authorized their use in or on "raw agricultural commodities" unless they were "unsafe" within the meaning of a newly enacted § 408.*fn21 That section provided that every pesticide chemical was "unsafe" unless exempted by the Secretary or used within a tolerance limit prescribed by the Secretary. Except insofar as courts would continue to read a de minimis exception into the new requirements,*fn22 and except insofar as the Administrator would decline to prosecute minor violations, every pesticide chemical was unsafe and caused adulteration unless and until it was the subject of a tolerance and thereafter used in conformity with the tolerance. As a matter of practice, a large number of tolerances were promulgated; the use of DDT and dieldrin was thereafter lawful as long as residues of these pesticides on raw agricultural commodities did not exceed specified limits.*fn23
Just as the pesticide chemical amendment in 1954 created a new category of added substances which automatically caused adulteration unless exempted or used in conformity with an established tolerance, so did the 1958 amendment create still another new category -- food additives -- which also, by definition, caused adulteration unless exempted or used within a prescribed tolerance. Section 402, describing adulterated food, was amended to include any food which bears or contains an "unsafe" food additive, and every such additive was deemed unsafe unless used in compliance with a tolerance regulation issued under a new § 409.
Thus, congressional efforts to define and regulate adulterated food in the 20 years before 1958 reflected a consistent desire to rationalize the law by reducing the circumstances in which the F.D.A. must prove actual danger to a quantity of food and by establishing a system of tolerances which will both protect the public and enable industry to operate effectively.
The legislative history of the 1958 Bill indicates concern about the difficulties present when dangerous substances could not be proscribed by per se rules.*fn24 Since Congress used broad language in order to eliminate such difficulties, we should not construe it narrowly. The language, set forth in full above,*fn25 defines a food additive as any substance, "the intended use of which results or may reasonably be expected to result, directly or indirectly, in its becoming a component or otherwise affecting the characteristics of any food . . . ." Although's Congress was primarily concerned with substances used by a food processor, neither the language nor the history of the 1958 Act limits its application to such substances.*fn26 The words "the intended use of which" are not confined, as they easily could have been confined, to use in food processing.
Vita has argued that a process, such as smoking, during which nothing new is added to a food, cannot "transmogrify" a preexisting component of a food into an additive. But whether the food be fish, fruit, or meat, if the component is a pesticide chemical, we think that is exactly what Congress intended.
Although it may seem odd to place the label "additive" on a chemical substance which was a component of the raw product and which is not changed by processing, Congress' choice of that label does not result in any "transmogrification." Before processing, DDT is a "pesticide chemical" on a raw product; after processing, it is an "additive." Since there is no tolerance for DDT on fish, both before and after processing the presence of the DDT causes fish to be adulterated without any proof that it is actually unfit as food. Defendants' contention, if accepted, would result in the anomaly that a chemical such as DDT would adulterate all raw fish, but adulteration of processed fish would be determined on an uncertain case-by-case basis. We conclude that such a construction of the statute is illogical and unacceptable.*fn27
This conclusion is confirmed by the full text of the description of products which are "unsafe" -- and therefore "adulterated" -- because they contain food additives. Subsection (a)(2)(C) of § 402 provides that a food is adulterated:
". . . if it is, or it bears or contains, any food additive which is unsafe within the meaning of section 409: Provided, That where a pesticide chemical has been used in or on a raw agricultural commodity in conformity with an exemption granted or a tolerance prescribed under section 408 and such raw agricultural commodity has been subjected to processing such as canning, cooking, freezing, dehydrating, or milling, the residue of such pesticide chemical remaining in or on such processed food shall, notwithstanding the provisions of sections 406 and 409, not be deemed unsafe if such residue in or on the raw agricultural commodity has been removed to the extent possible in good manufacturing practice and the concentration of such residue in the processed food when ready to eat is not greater than the tolerance prescribed for the raw agricultural commodity ;". 72 Stat. 1784. (Emphasis added.)
Quite clearly the need for the proviso arose from the fact that the definition of a "food additive" is broad enough to include any residue of a pesticide chemical remaining in or on food after processing. Prior to the 1958 amendment only raw agricultural commodities could be adulterated as a matter of law because of the presence of such an added chemical; processed foods were not subject to that risk. Accordingly, prior to 1958, the exemption for such pesticide chemicals used within tolerance limits on raw agricultural commodities was sufficient to avoid the risk of "per se" adulteration for all food subject to that risk. But when the "per se" concept was enlarged to encompass processed food, which might, of course, contain some chemical residue, it became necessary either (a) to make the "food additive" and the "pesticide chemical" categories mutually exclusive, or else (b) to enact a specific proviso to cover the area of overlap. Congress made the latter choice.
The proviso which Congress enacted avoids the necessity for duplicate tolerances, one covering the use of the chemical on the raw commodity and the second applying to the same chemical after processing. The fact that an express proviso was needed to avoid that consequence confirms our reading of the food additive definition as broad enough to encompass pesticide chemical residues in processed food.
Thus, the tolerances for DDT and dieldrin in or on raw fruits, vegetables, and meat are adequate to avoid adulteration caused by the residues remaining after the foods are processed. Without such tolerances, we think it is clear that the presence of DDT in or on such foods will cause adulteration pursuant to subparagraph (a)(2)(B) in their raw state, and that the same consequence will follow from subparagraph (a)(2)(C) after processing. We are also persuaded that these chemicals have the same impact on fish in the Great Lakes. As long as no tolerances have been established, the raw chubs are adulterated within the meaning of (a)(2)(B) because they contain an unsafe pesticide chemical; after processing, by virtue of (a)(2)(C), the chubs are adulterated because they contain an unsafe food additive. We think this is evident from the entire statutory scheme, the definitional language, and the relevant legislative history.*fn28 We hold, therefore, that the DDT and dieldrin found in defendants' smoked chubs were "food additives" and, since not protected by any tolerance, the chubs were "adulterated" as a matter of law.
Vita Food argues, in the alternative, that even if DDT and dieldrin are food additives, we should nevertheless affirm because the government failed to prove that residues of these chemicals exceeded the limits specified in the F.D.A. interim enforcement guidelines. In its opinion the district court stated that there is never "perfect certainty" in any scientific analysis and that the government's test methods were "not sufficiently reliable for me to find by the greater weight of the evidence and as a controlling fact that the chubs sampled by FDA in April 1972 contained DDT concentration in excess of 5 ppm." 356 F. Supp. 1213, 1221.*fn29
Neither party has contested the applicability of the F.D.A.'s 5 ppm enforcement guideline. Nevertheless, it is clear that the government might have argued that the enforcement guideline was merely adopted as an internal standard to determine when it would be appropriate to initiate an enforcement proceeding, and that publication in a Press Release was merely intended to give industry fair notice of its internal standards. By way of analogy, a police department might adopt a policy of not enforcing a 55 mile speed limit unless a motorist's speed exceeded 65 miles per hour. Under such a policy, at a trial it would not be necessary to prove anything more than a speed in excess of 55. By comparison, the F.D.A. might as a matter of discretion withhold enforcement unless it found residues of over 5 ppm, but have no legal obligation to prove anything more than a trace to establish statutory adulteration.*fn30
We do not so interpret the interim guidelines before us in this case; for both the language of the Press Release*fn31 and the government's treatment of it in this case*fn32 indicates that it was meant to operate more like a rule of general applicability than a mere prediction of how the agency would choose to exercise its prosecutorial discretion.*fn33 Accordingly, even though the government was not obligated to adopt any interim guideline, and might have let industry accept the responsibility for initiating tolerance proceedings, it correctly assumed the burden of proving that appellant violated the specified limits.*fn34
The district court considered the AOAC method of testing for chemical residues insufficiently precise for the government to sustain its burden. It is clear, however, that the enforcement guidelines must have been adopted on the assumption -- shared by government and industry -- that existing methods of testing DDT were sufficiently accurate to permit meaningful administration of the limits specified therein. The AOAC method was used by both the government technicians and by the expert employed by appellant. However imprecise that method may be, the record indicates that it is the best method that can be used. Certainly the government must be permitted to use the best testing method yet devised by analytical chemists, for the enforcement guidelines must have been predicated upon that method. Therefore, without disagreeing with the district court's observation that the AOAC method falls short of perfect certainty, we cannot accept the view that it may not be used to evaluate appellants' compliance with the guidelines. The district court's contrary determination was clearly erroneous.
Acceptance of the AOAC method as the proper standard for measuring residues of pesticide chemicals in fish leads inexorably to the conclusion that the government met its burden of proving repeated violations of the guidelines. In July and August of 1969, and again in the fall of 1969, the F.D.A. collected and analyzed smoked chubs shipped by Vita in interstate commerce and found quantities of DDT and its derivatives ranging from 5.97 ppm to 9.67 ppm. Tr. 663-665. After this case had been commenced, additional samples were analyzed and found to contain 5.37 to 9.28 ppm.*fn35 These samples were subsequently analyzed by an expert employed by appellant; he found that three out of the seven samples contained DDT in excess of 5 ppm and that the other ranged from 3.89 to 4.37. Finally, in October of 1972, samples were sent to Seattle for analysis with results ranging from 5.42 ppm to 6.54 ppm. Tr. 623.
Whether we accept the government's evidence at face value -- and since we have held that the AOAC procedure is acceptable, and since the detailed records supporting the government's conclusion are in the record, there is persuasive reason to do so -- whether we assume that the truth lies somewhere in between the views of the respective litigants' experts, or even if we take the defendants' evidence, it is perfectly clear that the limit specified in the 1969 guidelines has been violated. A shipper of processed food may violate the statute even if some, or even a majority, of its food is not adulterated. The evidence is uncontradicted that the Act, as implemented by the guideline, has been violated.
Vita argues, however, that there really is no significant difference between DDT levels of 5 ppm and levels as high as 8 ppm. But that is an argument that should not be addressed to us; it may properly be asserted as a reason for setting a tolerance at 8 ppm or perhaps even higher. The F.D.A. need not have set its guideline limit at 5 ppm, but it did so, and industry has not seen fit to invoke the statutory procedures for establishing a different tolerance level. In these circumstances, the government has met its burden by proving that the guidelines have been exceeded repeatedly.
The judgment in United States v. Vita Foods is therefore reversed and remanded for the entry of appropriate relief. In view of the nature of the case and the trial judge's familiarity with the technical materials in the record, Circuit Rule 23 shall not apply.
The judgment in United States v. Ewig Bros. Co. is affirmed.